POPCORN: Progressive Pseudo-labeling with Consistency Regularization and Neighboring
This work addresses the challenge of scarce annotated images and lack of generalization in medical segmentation tasks, though it appears incremental as it builds on existing SSL techniques.
The authors tackled the problem of medical image segmentation with limited labeled data by proposing POPCORN, a semi-supervised learning method combining consistency regularization and pseudo-labeling, which achieved competitive results on multiple sclerosis lesion segmentation compared to other state-of-the-art strategies.
Semi-supervised learning (SSL) uses unlabeled data to compensate for the scarcity of annotated images and the lack of method generalization to unseen domains, two usual problems in medical segmentation tasks. In this work, we propose POPCORN, a novel method combining consistency regularization and pseudo-labeling designed for image segmentation. The proposed framework uses high-level regularization to constrain our segmentation model to use similar latent features for images with similar segmentations. POPCORN estimates a proximity graph to select data from easiest ones to more difficult ones, in order to ensure accurate pseudo-labeling and to limit confirmation bias. Applied to multiple sclerosis lesion segmentation, our method demonstrates competitive results compared to other state-of-the-art SSL strategies.